> ## Documentation Index
> Fetch the complete documentation index at: https://docs.usetrident.dev/llms.txt
> Use this file to discover all available pages before exploring further.

# Trident: Cloud and AI Agent Security Platform

> Trident secures the cloud infrastructure your AI runs on and the agents you deploy to production — observability, red-teaming, and runtime firewall in one platform.

Trident gives security and engineering teams a single platform to observe, test, and protect AI agents in production while continuously mapping attack paths across cloud infrastructure. Whether you're tracing every prompt your agent makes, running automated red-team attacks before a release, or scanning your cloud for over-permissioned IAM roles that an agent could exploit, Trident connects those signals into one security graph.

<CardGroup cols={2}>
  <Card title="Quickstart" icon="rocket" href="/quickstart">
    Instrument your first agent and start receiving traces in under 5 minutes.
  </Card>

  <Card title="Core Concepts" icon="book" href="/core-concepts">
    Learn how agents, findings, the firewall, and the security graph work together.
  </Card>

  <Card title="Agent Security" icon="shield-halved" href="/agents/overview">
    Trace agents, run red-team attacks, and enforce runtime firewall policies.
  </Card>

  <Card title="Cloud Security" icon="cloud" href="/cloud/overview">
    Map cloud assets, discover attack paths, and generate remediation PRs.
  </Card>
</CardGroup>

## What Trident Does

Trident covers two interconnected attack surfaces that legacy tools treat separately: the cloud infrastructure your AI runs on and the AI agents that run inside it.

<CardGroup cols={3}>
  <Card title="Observe" icon="eye">
    Full-fidelity traces of every prompt, tool call, MCP invocation, and approval — with PII automatically redacted before leaving your process.
  </Card>

  <Card title="Test" icon="flask">
    200+ automated attack scenarios covering the OWASP Agentic Top-10. Run pre-deploy red-team campaigns from the dashboard or your CI pipeline.
  </Card>

  <Card title="Protect" icon="lock">
    Runtime firewall with two-stage prompt and output scanning. Confirmed findings automatically become ban rules enforced within 5 minutes.
  </Card>
</CardGroup>

## Get Started

<Steps>
  <Step title="Create a project">
    Sign in to [app.usetrident.dev](https://app.usetrident.dev) and create a new project. Copy your **Public Key** and **Secret Key** from the project settings.
  </Step>

  <Step title="Instrument your agent">
    Install the SDK for your language and call `trident.init()` (TypeScript) or `vouch_sdk.init()` (Python) once at startup. Traces start flowing immediately.
  </Step>

  <Step title="Review findings">
    Open the **Findings** inbox to see security issues discovered during tracing and red-team runs, each scored by OWASP and AIVSS severity.
  </Step>

  <Step title="Enable the firewall">
    Switch your agent's LLM base URL to the Trident gateway endpoint to route every call through the runtime firewall.
  </Step>
</Steps>

## Explore the Docs

<CardGroup cols={2}>
  <Card title="TypeScript SDK" icon="js" href="/sdk/typescript/installation">
    Install `@vouch-ai/sdk` and instrument Node.js agents with one function call.
  </Card>

  <Card title="Python SDK" icon="python" href="/sdk/python/installation">
    Install `vouch-sdk` to instrument Python agents, LangChain, CrewAI, and more.
  </Card>

  <Card title="REST API" icon="code" href="/api/authentication">
    Trigger red-team runs, query findings, and manage firewall rules programmatically.
  </Card>

  <Card title="CLI" icon="terminal" href="/cli/installation">
    Use `tridentctl` to scan your local environment and run evaluations from the command line.
  </Card>
</CardGroup>